Quicktime audio codecs, Sound sample rates – Apple Compressor 3 User Manual

Page 247

Advertising
background image

Lossy codecs: In comparison to lossless codecs, lossy codecs return only an approximation

of the data and are usually used to deliver finalized video to an end user. The level of
approximation is determined by the quality of the codec. Typically there is a tradeoff
between the output quality of the codec and its achievable level of compression. With
some codecs, a compression ratio of at least 5:1 can be reached without visible
degradation. An example of a lossy codec is the Photo-JPEG codec.

Low-quality lossy compression introduces unnatural visual artifacts that may become
noticeable to the human eye. Using a lossy codec, you should compress your media
file only once, because compressing a file multiple times adds more artifacts.

QuickTime Audio Codecs

If you have the disk space and bandwidth, the best thing you can do is leave your audio
uncompressed. Uncompressed audio normally uses 8-bit (phone quality) or 16-bit (CD
quality) samples. (Compressor supports up to 64 bits per sample floating point and a
maximum sample rate of 192 kHz.) If you have disk space and bandwidth limitations, you
must compress your audio. However, audio takes up less disk space than video, so it
doesn’t need to be compressed as much as video.

The two most important components of audio codecs are sample rates and size. The
sample rate sets the sound quality, and sample size sets the dynamic range of the sound.
QuickTime audio codecs allow you to set both the sample rate and size of your source
media file.

Sound Sample Rates

Digitized sound consists of sound samples captured at different frequency rates. The
more sound samples per second, the higher the sound quality. For example, audio CDs
use a 44.1 kHz sample rate, DVDs sample at 48 kHz, and telephone networks sample
voices at 8 kHz. The sample rate you choose depends on the nature of the sound. Music
requires a higher sample rate than voice, because music contains a wider range of
frequencies. Spoken voice has a more limited range of frequencies, so you can choose a
lower sample rate and still maintain acceptable audio quality. In most cases, you should
choose the highest sample rate available.

Reducing the sample rate can shrink a media file by as much as 5:1. The audio quality
will be affected, but not as much as it would be if you used 8-bit sampling. The following
table shows common sample rates and the audio device quality to expect at each rate:

Audio device quality

Sample rate

DAT/DV/DVD

48 kHz

CD

44 kHz

FM radio

22 kHz

Telephone

8 kHz

247

Chapter 20

Creating QuickTime Movie Output Files

Advertising